Markedly Reduced Expression of Annexin A2 Is Associated with Hypofibrinolysis and Provoked and Unprovoked Venous Thromboembolism

Introduction: Recent evidence indicates that plasma hypofibrinolysis, as measured by clot lysis times (CLT) at or above the 90^th percentile, can double the risk of venous thromboembolism (VTE). To our knowledge, there are no studies on the role of cell surface fibrinolysis in thrombosis. Annexin A2 (A2) is a calcium dependent, phospholipid binding protein that forms a heterotetramer with S100A10 (A2-S100A10)₂ on the surface of endothelial cells (ECs). This complex serves as a co-receptor for plasminogen and tissue plasminogen activator (tPA), and increases the catalytic efficiency of tPA-dependent cell surface plasmin generation by 60-fold. In mouse studies, global knockout of the annexin A2 gene (Anxa2) is associated with fibrin accumulation and impaired clearance of arterial thrombi. In addition, there are several examples of the regulatory role of A2 in fibrinolysis in human diseases such as antiphospholipid antibody syndrome, cerebral venous thrombosis, and sickle cell disease. In the current study, we aimed to explore the potential role of the A2 system and cell surface based fibrinolysis in the development of VTE.

Methods: Study subjects included patients 18-65 years old with history of VTE and healthy controls. Subjects were classified as having provoked or unprovoked VTE based on the presence or absence of identifiable environmental risk factors. Peripheral blood mononuclear cells (PBMCs) were isolated from whole blood collected in EDTA tubes, and used as a surrogate for ECs. Assays performed on PBMCs included determination of the rate of tPA-dependent cell surface plasmin generation using a fluorogenic substrate and analysis of CLT in the presence of phospholipid vesicles. Because A2 accounts for approximately 50% of cell surface based plasmin generation on both ECs and PBMCs, we analyzed total A2 protein expression relative to GAPDH in whole cell lysates of PBMCs using semi-quantitative western blotting. Additional assays included quantitative RT-PCR and ANXA2 gene sequencing. Statistical analysis was performed using one-way ANOVA followed by Tukey's multiple comparisons test.

Results: Overall, 116 subjects with VTE (76 with provoked and 40 with unprovoked VTE) and 87 healthy controls were studied between September 2010 and May 2019. Plasma based clot lysis assays revealed that the mean CLT for subjects with VTE was significantly longer than that for healthy controls (37.5 versus 30.7 minutes, p=0.0001). Additionally, the mean rate of cell surface plasmin generation was significantly reduced in subjects with VTE as compared with healthy controls (0.33 RFU/min² versus 0.49 RFU/min², p=0.0021). Moreover, none of the 60 healthy controls (0%), but 18 of the 107 subjects with thrombosis (17%) had cell surface plasmin generating capacity in the lowest 10^th percentile. In protein expression studies, we observed that none of the 21 healthy controls (0%), but 5 of 41 subjects with thrombosis (12%) had A2 expression in the lowest 10^th percentile for the group; 4 of 18 (22%) of those with unprovoked VTE and 1 of 23 (4%) of those with provoked VTE fell into the lowest decile for protein expression. For plasmin generating capacity, 8 of 36 (22%) of subjects with unprovoked VTE and 10 of 71 (14%) of those with provoked VTE occupied the lowest decile. Probing of western blots of samples obtained on two separate occasions several months apart with A2 epitope-specific antibodies revealed abnormalities within either the tPA binding N-terminal tail or C-terminal core domain in proximity to the plasminogen binding site. Neither quantitative RT-PCR nor ANXA2 gene sequencing of selected samples revealed abnormalities in either mRNA or genomic DNA that could explain the reduced A2 expression.

Conclusion: These data confirm findings previously reported by Lisman that plasma hypofibrinolysis is associated with VTE and may represent an independent risk factor for VTE. Additionally, we demonstrate for the first time that impaired cell surface based fibrinolysis and aberrations in A2 protein expression are associated with both provoked and unprovoked VTE, and may represent a novel risk factor for thrombosis. Possible explanations for reduced A2 expression include dysfunctional translation of mRNA into protein or post-translational proteolysis of the translated protein. Compromise of the A2-based fibrinolytic system may represent a previously unrecognized contributor to thrombophilia in VTE.